Accounting research, like many other social science disciplines, has gradually moved from qualitative to quantitative research with an emphasis on the use of multiple evidence or methods in the conduct of research. This chapter explores the concerns and implications of triangulation in the conduct of research in the social sciences, particularly in the field of accounting. Based on evidence from existing literature, the chapter submits that triangulation is an important strategy for enhancing the quality of accounting research. Accounting researchers, like those from other social science disciplines, often adopt triangulation when investigating a complex phenomenon whereby using a single data source or method may not allow an exhaustive investigation to fully understand it, hence the inability to reach a dependable conclusion. Despite the concerns and implications of use of triangulation in accounting and social science research, the chapter concluded it is a relevant approach especially at a time when adequate evidence and analytical rigor is required to substantiate research findings.
TopIntroduction
Validity and reliability are fundamental criteria upon which the quality of any research finding is adjudged. Validity is the extent to which a measure reflects the phenomena being studied and reliability is the extent to which studies are replicable (Ibiamke & Ajekwe, 2017). These two concepts mean different things in quantitative and qualitative research. In quantitative research, validity entails the extent to which an instrument actually measures what it intends to measure and how well it does measure it (Smith, 1990) and reliability concerns replicability or repeatability of results (Golafshani, 2003) while in qualitative research validity has to do with the quality of findings and reliability is about dependability of study findings (Golafshani, 2003; Bashir, Afzal & Azeem, 2008). Broadly, in qualitative research, validity and reliability concerns the issues of trustworthiness, dependability, credibility, transferability, and confirmability (Ibiamke & Ajekwe, 2017; Fusch, Fusch & Ness, 2018).
The qualitative nature of research which most social science researches takes poses problems of validity and reliability. Shenton (2004) for example pointed out that positivist researchers who operate on the belief that reality is ‘out there’ and exists independently of the researcher have often questioned the trustworthiness of qualitative research. The standardization of research design in the natural sciences and the selection of samples base on randomization according to the National Child Care Information Center (NCCIC, 2011) make these issues less problematic in pure sciences research than social sciences. Social science research designs are flexible and the selection of sample is more purposively done than drawn randomly (Shenton, 2004), therefore, tend to be associated with the problems of validity and reliability.
Social science researches which main element is human beings and involve the study of attitude, perception, feelings, behavior and other such qualitative attributes necessarily needs to meet these two criteria (i.e. validity and reliability). Generally, every research method, be it in terms of data collection or analysis has attendant strengths and weaknesses. Each data collection method including the questionnaire, interview, and observation as well as research designs like survey, case-study, experimental, historical design, etc or more broadly, qualitative and quantitative research, has its own weaknesses. Research findings and results are thus affected to some extent by the weaknesses associated with the research approach adopted by the researcher.
Over the years, to overcome the weaknesses associated with the use of particular or single research approach, a combination of approaches has been proposed. Such a strategy to research is known as “triangulation”. The Wikipedia encyclopedia (2011) explains triangulation to mean the use of more than two methods in a study with a view to double or triple check results. In other words, it is the utilization of multiple research methodologies by a researcher in the study of a phenomenon of interest. The objective of triangulation approach is to validate the findings of research as posited in the Wikipedia explanation. This view has however been contested by other scholars on the grounds of paradigm divided (see Burrell & Morgan, 1979). In particular, Burrell and Morgan (1979) argued that inter-paradigmatic journeys by researchers is rare. Notwithstanding, triangulation as a research strategy is stated to aim at enriching understanding of a phenomenon under examination (Hoque, Covaleski & Gooneratne, 2013) hence research quality.
This paper examines triangulation approaches as an evolving trend to research specifically within the domain of social science research. It specifically explores the meaning, forms, concerns and implications of triangulation to accounting research as well as proposes ways to resolve the contentious issues surrounding its use by accounting researchers. The paper is theoretical in nature therefore relies on previous published articles to deductively infer conclusion.